Cyberbullying Detection using NLP 💬

This project focuses on detecting and blocking bullying content in social media posts and comments using machine learning and natural language processing (NLP) techniques. Built with a Gradio web interface, it provides real-time monitoring to help prevent the spread of harmful content.

Live Demo 🚀

You can try the live application hosted on Hugging Face Spaces:

➡️ Try the Live App Here!

---

Table of Contents

1. Overview
2. Features
3. Installation
4. Usage
5. Model and Approach

---

Overview ℹ️

Cyberbullying on social media can have a significant impact on mental health. This project aims to create a safer online environment by identifying bullying content in real-time and blocking it before it reaches users. The application uses a trained machine learning model to classify content and alert moderators when bullying is detected.

---

Features ✨

• Real-time Detection: Automatically detects bullying content in posts and comments.
• NLP-based Analysis: Uses natural language processing to analyze the tone and intent of content.
• Simple Interface: Easy-to-use web interface for quick checks.

---

Installation 🛠️

To run the project locally, follow these steps:

1. Clone the repository:

   git clone [https://github.com/Sai2002Praneeth/cyberbullying-app.git](https://github.com/Sai2002Praneeth/cyberbullying-app.git)
   cd cyberbullying-app

2. Create and activate a virtual environment (for Windows):

   python -m venv .venv
   .\.venv\Scripts\Activate

3. Install the dependencies:

   pip install -r requirements.txt

4. Run the Gradio application:

   python app.py

5. Open the application in your browser at http://127.0.0.1:7860.

---

Usage 📝

1. Open the application in your browser.
2. Enter the social media post or comment text in the provided input field.
3. Click on Submit. The model will classify the content.
4. The result ("Bullying" or "Non-Bullying") will be displayed in the output box.

---

Model and Approach 🤖

The model was developed using machine learning and NLP techniques to analyze social media content. Key steps included:

• Data Collection: We compiled a dataset of over 40,000 social media posts and comments with labeled bullying content.
• Preprocessing: Text data was cleaned, tokenized, lemmatized, and stop words were removed.
• Feature Extraction: We extracted relevant features using the TF-IDF (Term Frequency-Inverse Document Frequency) technique.
• Model Training: Multiple classifiers were tested. The final model chosen was the Stochastic Gradient Descent (SGD) Classifier, which achieved an accuracy of 87% on the test set.

---

Analysis Repository 🔬

The full development process, including data exploration and model comparison, can be found in the cyberbullying-analysis repository.